DiffSHEG: A Diffusion-Based Approach for Real-Time Speech-driven Holistic 3D Expression and Gesture Generation

TL;DR

DiffSHEG introduces a diffusion-based model for real-time, synchronized 3D expression and gesture generation driven by speech, outperforming prior methods in quality and efficiency.

Contribution

It presents a novel diffusion-based transformer model for joint speech-driven expression and gesture generation with arbitrary length, including an outpainting sampling strategy.

Findings

Achieves state-of-the-art quantitative and qualitative results

Produces high-quality, synchronized 3D expressions and gestures

Validated by user study confirming superiority over prior approaches

Abstract

We propose DiffSHEG, a Diffusion-based approach for Speech-driven Holistic 3D Expression and Gesture generation with arbitrary length. While previous works focused on co-speech gesture or expression generation individually, the joint generation of synchronized expressions and gestures remains barely explored. To address this, our diffusion-based co-speech motion generation transformer enables uni-directional information flow from expression to gesture, facilitating improved matching of joint expression-gesture distributions. Furthermore, we introduce an outpainting-based sampling strategy for arbitrary long sequence generation in diffusion models, offering flexibility and computational efficiency. Our method provides a practical solution that produces high-quality synchronized expression and gesture generation driven by speech. Evaluated on two public datasets, our approach achieves state-of-the-art performance both quantitatively and qualitatively. Additionally, a user study confirms the superiority of DiffSHEG over prior approaches. By enabling the real-time generation of expressive and synchronized motions, DiffSHEG showcases its potential for various applications in the development of digital humans and embodied agents.